Human glioblastoma multiforme (GBM) responds poorly to currently available treatment. Other than surgery, radiation is the only method of treatment that significantly increases survival. However, conventional external beam photon radiation only prolongs survival and does not cure. Neutron radiation, which irradiates with high linear energy transfer (LET) radiation, has theoretical advantages over conventional low LET photon (x- ray) therapy, including: 1) a greater ability to damage hypoxic cells, 2) a lesser ability of tumor cells to repair high LET-induced damage, and 3) less variation in radiation sensitivity relative to the cell cycle. Preliminary work with external beam neutron therapy has shown that neutrons can completely eradicate GBM but that unacceptable radiation necrosis of normal brain frequently follows. The dose of radiation that can be given safely is limited by the tissue tolerance of the normal brain that surrounds the tumor. With external beam therapy, radiation must pass through normal tissue to reach the tumor. Interstitial radiation (from implanted radioactive sources) is a logical way to increase the dose to the tumor and decrease the dose to normal brain. Our preliminary work with Californium-252 (Cf-252), a neutron emitting element, has shown that Cf-252 has properties favorable for implantation and is feasible for treatment of GBM. The next step is to determine the optimum dose of radiation (the maximum tolerable dose (MTD)) for GBM therapy. A formal phase I trial (a dose escalation study) will be performed using 3 patients per dose step with each dose escalation of 100 neutron rads. Patients will have CT and MRI scans at regular intervals; if radiation necrosis is detected, a biopsy will be taken. The phase I study will be ended when 2 patients at a dose step develop biopsy proven radiation necrosis. The last dose step reached before necrosis will be the MTD. After the MTD is determined, an efficacy study (phase II study) will be performed. 50 patients with GBM will be give the MTD of neutron interstitial therapy and followed for at least one year after treatment. One year survival, overall survival, and time to recurrence will be endpoints. The results will be compared with historical controls to determine efficacy of CF-252 treatment. New advances in the technique of implantation using Cf-252 allow the use of large doses of qualitatively superior neutron radiation with reduced risk to normal tissue; Cf-252 therapy now promises a better chance of tumor control (or cure) than any currently available treatment for GBM.